Theiler's Virus-Induced Intrinsic Apoptosis in M1-D Macrophages Is ...

JOURNAL OF VIROLOGY, May 2008, p. 4502–4510 Vol. 82, No. 90022-538X/08/$08.00�0 doi:10.1128/JVI.02349-07Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Theiler’s Virus-Induced Intrinsic Apoptosis in M1-D Macrophages IsBax Mediated and Restricts Virus Infectivity: a Mechanism for

Persistence of a Cytolytic Virus�

Kyung-No Son,1,2 Robert P. Becker,3 Patricia Kallio,1 and Howard L. Lipton1,2*Departments of Neurology and Rehabilitation,1 Microbiology and Immunology,2 and Anatomy and Cell Biology,3

University of Illinois at Chicago, Chicago, Illinois 60612-7344

Received 30 October 2007/Accepted 13 February 2008

Theiler’s murine encephalomyelitis virus (TMEV), a member of the Cardiovirus genus in the family Picor-naviridae, is a highly cytolytic virus that produces necrotic death in rodent cells except for macrophages, whichundergo apoptosis. In the present study we have analyzed the kinetics of BeAn virus infection in M1-D cells,in order to temporally relate virus replication to the apoptotic signaling events. Apoptosis was associated withearly exponential virus growth from 1 to 12 h postinfection (p.i.); however, >80% of peak infectivity was lostby 16 to 24 h p.i. The pan-caspase inhibitor qVD-OPh led to significantly higher virus yields, while zVAD-fmkcompletely inhibited virus replication until 10 h p.i., precluding its assessment in apoptosis. In contrast, whilezVAD-fmk significantly inhibited BeAn virus replication in BHK-21 cells at 12 and 16 h p.i., virus replicationat these time points was not altered by qVD-OPh. Bax translocation into mitochondria, efflux of cytochrome cinto the cytoplasm, and activation of caspases 9 and 3 between �8 and 12 h p.i. (all hallmarks of the intrinsicapoptotic pathway) were transiently inhibited by expression of Bcl-2, which is not expressed in M1-D cells.Thus, BeAn virus infection in M1-D macrophages, which restricts virus replication, provides a potentialmechanism for modulating TMEV neurovirulence during persistence in the mouse central nervous system.

Theiler’s murine encephalomyelitis virus (TMEV), a mem-ber of the Cardiovirus genus in the family Picornaviridae, is ahighly cytolytic virus that produces necrotic death in rodentcells, including neurons and oligodendrocytes. The virus yieldin one-step growth kinetics studies in rodent cells is on theorder of 200 to 500 PFU/cell. An exception is infection ofmurine macrophages, which undergo programmed cell deathor apoptosis and produce virus yields restricted to �10 PFU/cell (10). The profile of TMEV-induced apoptosis and re-stricted virus yields has been observed during infection withlow-neurovirulence BeAn in a number of murine macrophagelines, including J774A.1, M1-D, P388D1, PU5-1.8, RAW264.7,IC21, and ANA-1 (11, 12), and in primary peritoneally derivedmacrophages (15). Analyses of another low-neurovirulenceTMEV, strain DA, have also revealed similar low virus yieldsduring infection in macrophages (26, 27, 29, 37). Murine pro-myelomonocytic (M1) cells in particular provide a useful invitro model to study the outcome of TMEV infection, sincethey can be induced to differentiate into mature macrophageswith interleukin-6 and then into activated macrophages withgamma interferon (IFN-�) (4, 35).

Limited data suggest that apoptosis following BeAn virusinfection is associated with caspase activation (11). IFN-�-activated M1-D cells to which BeAn virus has been adsorbedundergo apoptosis in the absence of detectable virus replica-tion (13), whereas UV-inactivated BeAn virus adsorbed to theM1-D cell surface does not induce apoptosis (15). IFN-� acti-

vation sensitizes these cells to death-inducing ligands, and virusinfection causes increased IFN-�/� secretion, resulting in up-regulation of TRAIL and tumor necrosis factor alpha ligandsthat mediate apoptosis (13).

Caspase-mediated apoptosis may be initiated through eitherof two broad pathways that are responsible for eventual celldeath. In the extrinsic pathway, which begins outside the cell,ligands bind to specific death receptors, such as Fas or thetumor necrosis factor receptor, in a conventional manner toactivate caspase-8, which in turn activates the executionercaspase, caspase-3 (21). When injury occurs within the cell, theintrinsic pathway is initiated at the mitochondrion by disrup-tion of the mitochondrial transmembrane potential and releaseof cytochrome c into the cytosol, where it binds Apaf1 toactivate caspase-9 and, in turn, activates caspase-3 (32). Incertain settings, caspase-8 activation may directly induce loss ofmitochondrial transmembrane potential through the activationof Bid. Mitochondrial homeostasis is regulated by both pro-and antiapoptotic Bcl-2 family proteins (7). The precise signal-ing pathways leading to apoptosis in M1-D cells during BeAninfection have remained unclear.

While persistence of noncytopathic RNA viruses, such aslymphocytic choriomeningitis virus, Borna disease virus, andhepatitis C virus, is readily understood, persistence of cytolyticRNA viruses, such as picornaviruses, is enigmatic, since con-tinuous cell-to-cell spread is required to perpetuate the infec-tion. Clearly, an RNA virus lytic for the target cell populationin which it persists provides no advantage for either the cellularreservoir or the host organism. Thus, either selection of atten-uated genetic variants in the viral quasispecies is required orhost factors associated with the target cell population itselfrestrict virus replication (23, 28). Since TMEV persisting inmice is not attenuated in neurovirulence (22), we examined the

* Corresponding author. Mailing address: Department of Microbi-ology-Immunology, MC 790, University of Illinois at Chicago, 835South Wolcott, Chicago, IL 60612-7344. Phone: (312) 996-5754. Fax:(312) 355-3581. E-mail: [email protected].

� Published ahead of print on 20 February 2008.

4502

on April 12, 2018 by guest

http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

possibility that apoptosis of infected macrophages, the princi-pal TMEV reservoir in the central nervous system (CNS) ofpersistently infected mice (1, 24, 30), and a population ofcontinuously replenished blood-borne monocytes crossing theblood-brain barrier to infiltrate demyelinating lesions (22a) isthe restricting element during persistent infection. Our analysiscorrelates the temporal kinetics of BeAn virus infection inM1-D cells with the hallmarks of the intrinsic apoptotic path-way in these cells.

MATERIALS AND METHODS

Cells and viruses. M1 cells, an immature myelomonocytic cell line derivedfrom the SL mouse strain, were induced to differentiate into macrophages withsupernatants from L929 and P388D1 cells as previously described (12). Theresulting M1-D cells were maintained in RPMI 1640 supplemented with 10%fetal bovine serum (FBS), 2 mM L-glutamine, and nonessential amino acids(complete medium). BHK-21 cells (purchased from the American Type CultureCollection) were maintained in Dulbecco’s minimum essential medium (Invitro-gen, Carlsbad, CA) containing 2 mM L-glutamine, 7.5% tryptose phosphate, and10% FBS. The origin and passage history of the BeAn virus stock have beendescribed elsewhere (34). Virus titers of clarified lysates of infected cells weredetermined by standard plaque assay in BHK-21 cells (34).

Virus infections. After virus adsorption at the indicated multiplicities of in-fection (MOIs) for 45 min at 24°C, cell monolayers were washed twice withphosphate-buffered saline (PBS) containing 1 mM CaCl2 and 0.5 mM MgCl2 andincubated in complete medium containing 1% FBS at 37°C for the times indi-cated.

Reagents. Pan-caspase inhibitors zVAD-fmk, qVD-OPh, and rabbit anti-caspase-8 were purchased from R&D Systems (Minneapolis, MN); mouse anti-caspase-9, rabbit anti-caspase-3, rabbit anti-poly(ADP-ribose) polymerase(PARP), rabbit anti-Bak, rabbit anti-Bcl-2, rabbit anti-actin, and rabbit anti-cytochrome c antibodies were from Cell Signaling Technology (Beverly, MA);mouse anti-Bax and small interfering RNA (siRNA) to Bax were obtained fromSanta Cruz Biotechnology Inc. (Santa Cruz, CA); mouse anti-cytochrome c wasfrom Calbiochem (Darmstadt, Germany); goat anti-mouse immunoglobulin G(IgG)–horseradish peroxidase (HRP), goat anti-rabbit IgG, and goat anti-mouseIgG–fluorescein isothiocyanate were from BD Pharmingen (San Diego, CA);goat anti-rabbit IgG was from Abcam (Cambridge, MA); enhanced chemilumi-nescence solution was purchased from Amersham (Piscataway, NJ). pcDNA3/Bcl-2, pcDNA3/Bcl-xL, and pcDNA3 were kindly provided by Xuming Zhang,Little Rock, AR.

Cell viability assay. The tetrazolium salt WST-1 (Roche Applied Science,Indianapolis, IN) was added to the medium of monolayer cultures in 35-mmmultiwell dishes at the indicated times and incubated for 1 to 2 h at 37°C in a 5%CO2 atmosphere. Cell viability was determined by absorbance at 420 nm (ref-erence wavelength, 610 nm) using a microplate (ELISA) reader for cleavage ofthe tetrazolium salt to formazan against a background control. Values werecalculated as the ratio of cell death in BeAn virus-infected cultures to that inmock-infected cultures.

TdT-mediated dUTP-biotin nick end labeling (TUNEL) assay. M1-D cells,grown and infected on glass coverslips (Fisher Scientific Co., Pittsburgh, PA),were fixed in 85% ethanol–15% acetone and permeabilized with PBS containing0.1% Triton X-100 for 3 min to detect apoptotic cells using a FlowTACS detec-tion kit (R&D Systems) according to the manufacturer’s instructions. Briefly,infected cells on coverslips were washed twice in distilled water for 2 min,transferred into 1� terminal deoxynucleotidyltransferase (TdT) labeling bufferfor 1 min, and incubated with the labeling reaction mix (50 �l/coverslip) for 1 h.Coverslips were transferred into terminal TdT stop buffer for 5 min, washedtwice in PBS, and incubated with 50 �l of Strep-Fluor (1:200) for 20 min and with0.5 �g/ml 4,6-diamidino-3-phenylindole solution in PBS for 3 min. Coverslipswere washed with 0.5% Tween 20 in PBS and distilled water and viewed with aZeiss digital confocal microscope.

Subcellular fractionation. A subcellular proteome extraction kit (Calbiochem)was used to isolate cytosol and heavy mitochondria membrane fractions fromM1-D cells according to the manufacturer’s instructions. M1-D monolayers at adensity of 1 � 106 cells/35-mm well were washed twice and incubated for 10 minat 4°C with gentle agitation in ice-cold extraction buffer I containing proteaseinhibitors. Removal of the supernatant provided the cytosolic fraction. Mono-layers incubated for 30 min at 4°C with gentle agitation in extraction buffer IIcontaining protease inhibitors were the source of the heavy membrane fraction.

Fluorescence-activated cell sorter (FACS) analysis of cytochrome c. The In-nocyte kit (Calbiochem) for cytochrome c release from mitochondria that relieson selective permeabilization of cell membranes for release of cytosolic compo-nents leaving mitochondrial membranes intact was used following the manufac-turer’s instructions. Cytochrome c was detected with a mouse monoclonal anti-body and a fluorescein isothiocyanate-conjugated secondary antibody by flowcytometry.

Immunoblot analysis of cellular proteins. M1-D cells (monolayer cells andcells shed into the supernatant combined) were washed with PBS and lysed inradioimmunoprecipitation assay buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl,1 mM EDTA, 1% NP-30, 0.5% sodium deoxycholate, and 0.1% sodium dodecylsulfate) at the indicated times. Protein samples were electrophoresed on 12%NUPage bis-Tris gels (Invitrogen) and transferred to nitrocellulose membranes.Membranes were blocked with Tris-buffered saline containing 3% nonfat drymilk and 0.02% Tween 20 and incubated with primary antibody for 1 h and with1:100 goat anti-rabbit–HRP or anti-mouse–HRP as the secondary antibodies for1 h. A rabbit antibody to �-actin (Cell Signaling) was used as a loading control.Antibody dilutions were determined from initial experiments with M1-D cellsinduced to undergo apoptosis by treatment with 1 �g/ml actinomycin D. Quan-tification of the immunoblots was performed with the program Prism 4 (Graph-Pad Software, San Diego, CA).

Immunofluorescence staining. M1-D cells, grown and infected on glass cov-erslips, were fixed and permeabilized (see “TUNEL assay,” above). Infected cellswere washed twice in PBS, incubated with 1:1,000 rabbit anti-BeAn serum(detects capsid proteins and immediate precursor), 1:100 rabbit anti-cytochromec (Cell Signaling), or mouse anti-Bax monoclonal antibody for 30 min, washedonce in PBS, and incubated with 1:200 goat anti-rabbit IgG (Abcam) or 1:200goat anti-mouse IgG (BD Pharmingen) for 30 min. Coverslips were inverted onmicroscope slides onto gel mount (Biomedia, Foster City, CA) and viewed witha Zeiss digital confocal microscope.

Microscopy. M1-D cells were harvested and fixed with 3% glutaraldehyde inPBS. Cell were further fixed in aqueous 2% osmium tetroxide, stained with 0.5%aqueous uranyl acetate, dehydrated with a graded ethanol series, and embeddedin epoxy resin LX112. Transverse sections (1 �m) were cut and further stainedwith toluidine blue O for light microscopy. For transmission electron microscopy,sections were cut at a 100-nm thickness, placed on Formvar-coated 200-meshcopper grids, stained further with uranyl acetate and lead citrate, and viewedunder a JEOL model 1220 microscope (Tokyo, Japan) at 80 kV and with 1,000�to 150,000� magnification. Images were documented with a Gatan multiscancamera model 794.

Statistical analysis. A paired Student’s t test was used to compare groups, anddifferences were considered significant at a P level of �0.05.

RESULTS

Temporal kinetics of apoptosis in BeAn virus-infected M1-Dcells. Death of M1-D cells infected with BeAn virus (MOI, 10)was apparent at 8 to 10 h postinfection (p.i.) and increasedthereafter, as determined by cell survival assay (Fig. 1A). Phasemicroscopy revealed subtle signs of cytopathology at 8 h p.i.,with increasing numbers of pyknotic and fragmented cells seenat 12 to 16 h p.i. (Fig. 1B). A TUNEL assay of infected cellmonolayers showed stained cells as early as 6 h p.i. (Fig. 1B).Electron microscopy identified autophagy in monolayer cells at12 and 16 h p.i., with typical features of double-layered mem-brane vacuoles and arcuate cytoplasmic clefts (Fig. 1C, cell b),whereas detached cells exhibited features of apoptosis, includ-ing condensed nuclei and nuclear chromatin and cytoplasmicblebbing (Fig. 1C, cell c). Some apoptotic cells contained au-tophagic vacuoles, and apoptotic fragments of cells were alsoobserved in the supernatant fraction (Fig. 1C, cell d). Perinu-clear collections of proliferative vesicles, indicative of sites ofvirus replication, were seen in some apoptotic cells (Fig. 1C,cell c) (36). Thus, both TUNEL and electron microscopic anal-yses indicated that M1-D cells undergo apoptosis as a result ofBeAn virus infection.

VOL. 82, 2008 TMEV-INDUCED APOPTOSIS IN M1-D MACROPHAGES 4503


http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

Decrease in BeAn infectious virus after 12 h p.i. in associ-ation with apoptosis of infected M1-D cells. To correlate theobserved temporal changes in apoptosis with the infection,virus antigens were detected by immunofluorescent stainingand virus yields were measured by standard plaque assay (Fig.2). Virus antigen expression was detected at 2 h p.i. (not seenin this image reproduction) with increasing numbers and in-tensity of staining between 4 and 12 h p.i. (Fig. 2A). Previously,BeAn virus yields in M1-D cells were only assessed at 24 h p.i.(14) and suggested that virus replication may have been re-stricted throughout the infection. Our present analysis re-vealed increasing virus yields to 12 h p.i., similar to that inBHK-21 cells (not shown), but with two- to threefold-loweramounts of virus than in BHK-21 cells at each time point.Unlike the infection in BHK-21 cells, virus yields in M1-D cellssteadily declined after 12 h p.i. (Fig. 2B and C). A similarkinetics but with high virus yields to 24 h has also been seen inL929 cells which, unlike BHK-21 cells, are not deficient in IFNproduction (unpublished data). However, addition of poly-clonal anti-IFN antibodies to the medium of BeAn virus-in-fected M1-D cells after virus adsorption did not increase virus

yields after 12 h p.i. (not shown), suggesting that apoptosis, notIFN activity, restricted virus yields. Addition of zVAD-fmk toBeAn virus-infected BHK-21 cells also significantly reducedvirus at 12 and 16 h p.i. but not at 20 h p.i., whereas additionof qVD-OPh did not increase virus yields as it did in M1-Dcells (Fig. 2C).

Effects of pan-caspase inhibitors on BeAn virus infection.Based on previous analysis suggesting that BeAn virus-inducedM1-D cell death is caspase mediated (12), we examined theeffect of the pan-caspase inhibitor zVAD-fmk on infection.Although zVAD-fmk (20 �M) added to the medium inhibitedvirus replication until 12 h p.i., limited virus replication wasobserved at later times (Fig. 2B). zVAD-fmk has been re-ported to inhibit 2A of human rhinoviruses (5). Since cardio-virus 2A does not have proteolytic activity, it is likely thatzVAD-fmk inhibits the only other cardiovirus protease, 3Cpro

(21a). Thus, the prevention of caspase-3 and PARP cleavage at12 h p.i. and protection of M1-D cells at 16 h p.i. (P � 0.001)by zVAD-fmk (Fig. 3A and B) was most likely due to its knownantiviral activity rather than as a pan-caspase inhibitor. Be-cause the antiviral effect of zVAD-fmk was transient, slight

FIG. 1. Temporal profile of cell death and morphology of BeAn virus-infected M1-D cells (MOI, 10) undergoing apoptosis. A. Cell survival plotshowing onset of cell death at 8 to 10 h p.i., with progressive cell death thereafter (means standard deviations). B. Morphology of TUNEL-positive cells at 6 to 8 h p.i.; note the small shrunken cells at 10 to 16 h p.i. revealed by phase-contrast microscopy. C. Electron microscopy profileof infected cells: (a) mock-infected cell, showing organelles including granules (�) reminiscent of circulating monocytes. Bar, 2 mm; (b) autophagiccells in the monolayer, showing double membrane phagocytic vacuoles (a�) and arcuate cytoplasmic clefts (�) at 12 h p.i.; (c) floating apoptoticcell with nuclear chromatin condensation and perinuclear bodies of proliferative vesicles (pv) at 16 h p.i.; (d) floating apoptotic fragments of cellsat 16 h p.i.

4504 SON ET AL. J. VIROL.


http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

cleavage of these products was seen at 16 h p.i. (Fig. 3A). Incontrast, 10 �M qVD-OPh did not inhibit early virus replica-tion but partially prevented caspase-3 and PARP cleavage at12 and 16 h p.i. (Fig. 3A) (P � 0.001), protected cells fromdeath until 16 h p.i. (Fig. 3B) (P 0.005), and resulted in asignificant increase (P � 0.01) in virus yields at 12 h and 16 hp.i. (Fig. 2B). The protection by qVD-OPh was no longer seenat 24 h p.i., and higher concentrations (20 and 50 �M) did notincrease cell survival time (not shown). qVD-OPh did not have

a similar protective effect in infected BHK-21 cells which un-dergo necrosis and not apoptosis (Fig. 2C). The antiviral effectof zVAD-fmk is due to the carboxy-terminal fluoromethylk-etone moiety, which binds the active proteolytic site, whereasqVD-OPh has a carboxy-terminal phenoxy group conjugatedto Val and Asp residues (2). Thus, BeAn virus infection re-sulted in caspase-mediated apoptosis that was associated withloss of infectivity after 12 h p.i., providing a potential mecha-nism of attenuating this lytic picornavirus during its CNS rep-lication in macrophages in the mouse.

Caspase proteolytic activity implicates the intrinsic apop-totic pathway. The temporal pattern of signal induction of theinitiator and effector caspases and PARP was examined at 4-hintervals postinfection by immunoblotting of cell lysates.Cleavage of PARP was observed at 8 h p.i., caspase-9 and -3 at12 h p.i., and caspase-8 to its fully active p18 form at 16 h p.i.,followed by increasing cleavage of pro- to active caspase formsat subsequent times (Fig. 4A). Levels of pro- and activecaspase forms decreased after 12 h p.i., due to increasingprotein degradation and cell death (Fig. 4A, loading control�-actin). Figure 4B shows the results of densitometric analysisof immunoblot autoradiograms to quantitate the kinetic profile

FIG. 2. Temporal kinetics of BeAn virus replication in M1-D andBHK-21 cells. A. Immunofluorescence antibody staining of BeAn virusantigens, revealing cytoplasmic virus antigens in all cells by 4 to 6 h p.i.Note the numerous shrunken cells at 12 to 16 h p.i. B. Infectious virusyields (solid line) at the indicated times postinfection, revealing expo-nential virus growth until 12 h p.i. and precipitous loss of infectivitythereafter (means standard deviations). Virus growth was inhibitedafter treatment with the pan-caspase inhibitor zVAD-fmk (dotted line)but increased after treatment with the pan-caspase inhibitor qVD-OPh(dashed line). C. BeAn virus titers at 12 to 24 h p.i. in BHK-21 cells,showing a significant decrease (P 0.03) in the titer at 12 and 16 h p.i.with the pan-caspase inhibitor zVAD-fmk (�) but no change in titer withqVD-OPh present in the medium (means standard errors).

FIG. 3. Effects of pan-caspase inhibitors on BeAn virus-inducedapoptosis, including caspase-3 cleavage to its 17-kDa active form,PARP-1 cleavage, and cell protection of BeAn virus-infected M1-Dcells. A. zVAD-fmk and qVD-OPh markedly inhibited caspase-3 cleav-age to the active form (p17) and PARP-1 cleavage at 12 h p.i., withpartial inhibition at 16 h p.i. B. Significant protection against cell deathwas provided by zVAD-fmk (P 0.001) and qVD-OPh (P 0.005) at16 h p.i.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

from the ratio of the cleaved form to proenzyme form. Neithercaspase-9 nor caspase-3 was cleaved at 10 h p.i. (data notshown). Caspase-9 does not require cleavage to become active;instead, caspase-9 is activated by dimerization (9, 33). Thus, itis likely that caspase-9 activation as well as caspase-3 cleavageand activation occurred earlier than 12 h p.i. and accounted forPARP cleavage and cytochrome c efflux into the cytoplasm at8 h p.i. (see below). Moreover, the caspase-3 antibody usedappears to be more sensitive in detecting the proform than theactive cleavage product, possibly accounting for the delayedappearance of caspase-3 cleavage. As shown in Fig. 4A,caspase-8 was already partially cleaved to p41 at 0 h (after virusadsorption). Overall, the proteolytic profile is consistent withan intrinsic apoptotic pathway underlying death of BeAn-in-fected M1-D cells. Cleavage of Bid to its active form tBid onlyat 16 h p.i. provided further evidence against involvement ofthe extrinsic apoptotic pathway (Fig. 4C and D).

Cytochrome c release from mitochondria. Activation of theinitiator caspase-9 results from freeing of proapoptogenic cy-tochrome c from mitochondria. Analysis of cytochrome c effluxinto the cytosol as a surrogate indicator of mitochondrial outermembrane permeabilization in BeAn virus-infected M1-Dcells by indirect immunofluorescence staining of infected cells

revealed cytochrome c in mitochondria based on particulatestaining until 6 h p.i. (4 h p.i. data not shown), whereas morediffuse staining throughout the cytoplasm was first seen at 8 hp.i. and at 10 and 12 h p.i. (Fig. 5A). An increase in cytosoliccytochrome c was first seen at 8 h p.i. as a distinct cap ofstaining around the nucleus (Fig. 5A). This pattern of cyto-chrome c efflux into the cytoplasm was also demonstrated byFACS analysis, which showed its release at 8 h p.i. (Fig. 5B).Moreover, immunoblotting of cytosolic and heavy membranefractions separated from whole-cell extracts showed reactivitywith antibodies to cytochrome c in the cytosol at 8 h, withincreasing amounts released at 10 to 16 h p.i. (Fig. 5C and D).Thus, the kinetics of cytochrome c efflux into the cytoplasmfrom mitochondria paralleled that of caspase-9 activation, con-sistent with caspase-9 activation from dimerization prior to itscleavage observed at 12 h p.i.

Infection results in Bax translocation to the mitochondria.Analysis of the immediate upstream signals responsible forcytochrome c release in BeAn virus-infected M1-D cells re-vealed an increase in Bax, but not Bak, expression that wasseen at 4 h p.i., increased thereafter, reaching fivefold greaterlevels at 10 h p.i. than in cells at the end of adsorption (0 h p.i.)(Fig. 6A). This result suggests that Bax expression is inducedby the infection. Bax is located in the cytoplasm of viable cellsand upon activation translocates to the mitochondria and isinserted into the mitochondrial outer membrane (8, 40). Im-munoblotting to detect Bax translocation to the mitochondriain infected fractions using a mouse monoclonal antibody toBax indicated decreased amounts of Bax in the cytosol duringtranslocation into mitochondria in infected cells at 8 to 10 h p.i.(Fig. 6B and C), followed by a further decrease in Bax cytosoliclevels at 12 to 16 h p.i. (Fig. 6B). Immunofluorescence stainingof infected cells also revealed a shift from diffuse to particulateBax cytoplasmic staining in a portion of cells at 8 h p.i. and inall cells at 10 h p.i. (Fig. 6D). Together the results indicate theessentially concomitant translocation of Bax and the release ofcytochrome c from the mitochondria, correlating with the ac-tivation of caspase-9.

Bax inhibition by RNAi abrogates cytochrome c release frommitochondria. To confirm the requirement of Bax in mito-chondrial permeability, Bax expression was knocked down byRNA interference (RNAi) in BeAn virus-infected M1-D cells.Immunoblot analysis of cells transfected with Bax siRNA (CellSignaling) showed reduction of Bax to 25% the level in in-fected cells at 10 h p.i. compared to cells transfected with anirrelevant siRNA (Fig. 7A). Cytochrome c release was assessedin both transfected cell populations and revealed a 66% reduc-tion in cytochrome c release into the cytosol in cells transfectedwith Bax siRNA compared to an irrelevant siRNA (Fig. 7A).In addition, transfection of Bax siRNA in infected M1-D cellsled to modest but significant (P 0.01) cell survival at 16 h p.i.(Fig. 7B). These results demonstrate that Bax is required formitochondrial permeability and cytochrome c release and theinduction of apoptosis in BeAn-infected M1-D cells.

Protection of infected M1-D cells by Bcl-2 but not Bcl-xL.The antiapoptotic Bcl-2 family proteins Bcl-2 and Bcl-xL playa central role in inhibiting the mitochondria-dependent celldeath pathway. Moreover, it has been reported that Bcl-2 butnot Bcl-xL is expressed in promyelomonocytic M1 cells,whereas differentiation of M1 cells into M1-D macrophages

FIG. 4. Immunoblot analysis of caspase proteolytic activity indi-cates activation through the intrinsic pathway. (A) Cleavage ofPARP-1 was observed at 8 h p.i. and of caspase-9 and caspase-3 at 12 hp.i., with increasing caspase and PARP-1 cleavage occurring at subse-quent times. Protein degradation due to cell death was seen after 12 hp.i. (see �-actin loading control). (B) Densitometric analysis of auto-radiograms of panel A. (C and D) Immunoblot and densitometricanalysis, respectively, of BID cleaved to its active form tBID at 16 h p.i.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

results in loss of Bcl-2 and upregulation of Bcl-xL expression(11), as recapitulated in Fig. 8A. Analysis to test whether Baxis held in check by an antiapoptotic Bcl-2 family protein(s)using M1-D cells transfected with pcDNA3-Bcl-2 and pcDNA340 h before infection (transfection efficiencies of 35 to 40%were obtained in this macrophage cell line cotransfected withpEGFP-N1) showed that Bcl-2 resulted in inhibition of cleav-age of caspase-9 and caspase-3 to their active forms at 12 h p.i.but not at 8 h p.i. (Fig. 8B and C). Cleavage of these caspasesearlier than in untreated cells (Fig. 4) may be due to cell stressfrom transfection. Although caspase inhibition causes a shift tocaspase-independent self-destructive processes and often doesnot prevent cell death, particularly in macrophages (20, 37a),41 0.81% (mean standard deviation) of M1-D cells trans-fected with pcDNA3-Bcl-2 before infection survived to 16 hp.i., compared to 32 2.56% of cells transfected with pcDNA3(P 0.03; n 3) (Fig. 8D). However, this modest but signif-icant (P 0.03) protection of cell monolayers was lost at 20 hp.i. (not shown). In contrast, overexpression of Bcl-xL did notprevent caspase-9 and caspase-3 cleavage or protect from celldeath (not shown). These data further support the conclusionthat BeAn virus-induced apoptosis involves the intrinsic path-way mediated by Bax.

DISCUSSION

In the present study we examined the kinetics of BeAn virusreplication in M1-D cells with respect to apoptotic signaling

FIG. 5. BeAn virus infection leads to mitochondrial outer membrane permeabilization and cytochrome c release. A. Immunofluorescenceantibody staining revealed particulate staining until 8 h p.i., when more diffuse staining of the cytoplasm was observed. This latter pattern appearedas a polar cap partially surrounding the nucleus. B. FACS analysis showed release of cytochrome c into the cytoplasm (dashed line) at 8 h p.i., withincreasing amounts of cytochrome c subsequently released. C. Immunoblots of whole-cell extracts separated into cytosolic and membrane fractionsrevealed slight cytochrome c release into the cytosol at 8 h p.i., which increased at 10 and 12 h p.i. and diminished at 16 h p.i., concomitant withincreasing cell death. D. Densitometric quantitation of the ratio of cytochrome c in the autoradiograms of the immunoblot of panel C.

FIG. 6. Infection of M1-D cells results in Bax translocation fromthe cytoplasm to mitochondria. (A) Immunoblot showing increasedBax expression in the cytosol at 8 to 10 h p.i., while Bak levels remainthe same. (B and C) Immunoblot and densitometric analyses, respec-tively, of subcellular fractions from infected M1-D cells revealingtranslocation of Bax from the cytosol to mitochondria at 8 to 10 h p.i.(D) Fluorescence staining of infected M1-D cells, revealing a shiftfrom a diffuse or cytoplasmic pattern to a particulate or mitochondrialpattern at 8 to 10 h p.i. (numbers indicate the hours postinfection).



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

events in these cells. Previous evidence that BeAn virus-in-duced apoptosis involves caspase activation (11) led us to an-alyze the apoptotic pathway(s) involved. Electron microscopy,the gold standard for demonstrating apoptosis, revealed dou-ble membrane structures typical of autophagosomes before theappearance of the classic apoptotic features of cellular andnuclear shrinkage, chromatin condensation, blebbing, nuclearfragmentation, and fragmentation of apoptotic bodies in in-fected M1-D cells (16). Classifications of apoptosis include anautophagic type (type 2), and autophagy, a homeostatic pro-cess related to apoptosis, which provides a system to removedamaged organelles and aggregated proteins, functions as acell survival mechanism (18). Blocking of autophagy leads toincreased susceptibility to proapoptotic insults (31).

The many cellular stresses that can trigger autophagy in-clude infections, and recent studies recognize that mammaliancells infected with positive-strand RNA viruses accumulatedouble membrane vesicles of autophagic origin which are be-lieved to serve as a physical scaffold for virus replication com-plexes (viroplasm) (reviewed in references 17 and 38). Themorphological events shown here in BeAn virus-infectedM1-D cells might represent a failure of the autophagic systemin cell survival, culminating in infected cells that progress toapoptosis.

Exponential virus replication typical of one-step growth ki-netics of most picornaviruses was observed between 1 and 12 hp.i. in infected M1-D cells; however, �80% of peak infectivitywas lost by 16 to 24 h p.i. (Fig. 2B). Previous studies of TMEV-induced apoptosis indicated that a higher MOI restricts final

virus yields by 99% (10, 11, 14). In contrast, TMEV replicationin rodent cell lines such as BHK-21 cells results in necrosis,with no reduction of peak infectivity titers at later times (Fig.2C). These observations suggest that the loss of infectivity isdue to apoptosis. Our preliminary data indicate that caspaseactivation associated with TMEV one-step growth kinetics inM1-D cells leads to cleavage of VP2 in 150S virions (fromsucrose gradients) at 16 h p.i. but not earlier, resulting in highparticle-to-PFU ratios.

The present study also shows that BeAn virus infection in-duces apoptosis through the intrinsic pathway. Hallmarks ofactivation of the intrinsic apoptotic pathway were observed at8 to 12 h p.i. during exponential BeAn virus growth in M1-Dcells. Bax translocation to the mitochondria outer membraneand release of cytochrome c into the cytoplasm occurred at 8 to10 h p.i. (Fig. 5) when caspase-9 had probably dimerized andbecome activated (9, 33), with caspase-9 cleavage followinglater at 12 h p.i. (Fig. 4A and B). PARP cleavage and positiveTUNEL staining at 8 h p.i. (Fig. 1C and 4A and B) suggest thatat least some caspase-3 was cleaved to the active form beforethis was demonstrated on immunoblot assays. This interpreta-tion of the temporal signaling events is also consistent with thelinear cell death observed at 10 to 16 h p.i. (Fig. 1A). Partialcleavage of capase-8 to p45 was seen in uninfected and mock-infected M1-D cells. Caspase-8 cleavage to the active p18 formand cleavage of Bid to tBid occurred at 16 h p.i. and most likelywas due to activated caspase-3 “feeding back” to cleave

FIG. 7. Bax is required for cytochrome c release from mitochon-dria. A. Immunoblotting revealed that transfection of Bax siRNA ininfected M1-D cells reduced Bax expression at 10 h p.i. to 25% of thelevel in cells transfected with an irrelevant siRNA and reduced cyto-chrome c release from mitochondria by 66%. B. Transfection of BaxsiRNA in infected M1-D cells led to modest but significant (P 0.01)cell survival at 16 h p.i.

FIG. 8. Expression of Bcl-2 protects against BeAn virus-inducedapoptosis. A. Immunoblot showing that Bcl-xL, but not Bcl-2, is ex-pressed in M1-D cells. B. Immunoblot showing inhibition of caspase-9and -3 cleavage to their active forms at 12 h p.i. in cells transientlytransfected with pcDNA3/Bcl-2 but not in cells transfected withpcDNA3. C. Densitometric analysis of the ratio of active to pro- oractive forms of the caspases in panel B. D. Cell survival assay demon-strating significant protection (P 0.03) from cell death by expressionof Bcl-2 compared to vector only.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

caspase-8, arguing against a role for the extrinsic apoptoticpathway. Moreover, the inability of UV-inactivated TMEV toinduce apoptosis in mammalian cells (14) precludes signalingfrom engagement of death-inducing ligands on the cell surface.

The antiapoptotic Bcl-2 family plays a central role in regu-lating apoptosis in immune cells (25). Our analyses demon-strate that expression of Bcl-2, not normally expressed inM1-D cells (Fig. 7A), inhibits caspase-9 and caspase-3 cleavageand provides significant, albeit modest, cell protection; how-ever, overexpression of Bcl-xL had no rescue effect (notshown) for unexplained reasons. The antiapoptotic Bcl-2 fam-ily member Mcl-1, initially isolated from a human myeloblasticleukemia cell line (19), reportedly predominates in differenti-ated human macrophages (25) and protects against apoptosisduring the initial steps of differentiation (3, 41). Mcl-1 has alsobeen described as essential for differentiation and survival ofmacrophages (3, 41); however, Dzhagalov et al. (6) generatedMcl-1 conditional knockout mice in which endogenous expres-sion of Mcl-1 was required for neutrophil but not macrophagesurvival. Nonetheless, Mcl-1 is likely to be a key antiapoptoticregulator in macrophages, and its overexpression in M1-D cellsmay not only provide greater protection against BeAn virus-induced apoptosis but also may point to specific upstreamproapoptotic BH3-only activators, e.g., Nova, Puma, and Bim,which sense cellular damage and engage their antiapoptoticrelatives to overcome the block to Bax or Bak activation (39).

Our analyses were conducted using a cell type (macro-phages) relevant to the persistence of low-neurovirulenceTMEV in the mouse CNS, in contrast with the many reports onother animal viruses in which apoptosis has been induced incells that bear little or no resemblance to the infected cellpopulation(s) and pathogenetic events in vivo. The fact thatTMEV infection in other rodent cell lines produces necrosisand not apoptosis underscores the importance of the host cellfor downstream cellular events during infection. TMEV infec-tion in primary macrophages also induces apoptosis (15), sup-porting the relevance of the infection in M1-D cells in vitro.Thus, cytolytic BeAn virus infection in M1-D cells and re-stricted virus replication provide a potential mechanism forreducing TMEV neurovirulence of this virus during persis-tence in the mouse CNS and allowing host cell survival.

ACKNOWLEDGMENTS

We thank Prasad Kanteti for helpful discussions and Marina Hoff-man for editorial assistance.

This work was supported by NIH grant NS23349, the ModestusBauer Foundation, and the Wershkoff Multiple Sclerosis ResearchFund.

REFERENCES

1. Aubert, C., M. Chamorro, and M. Brahic. 1987. Identification of Theiler’svirus infected cells in the central nervous system of the mouse during demy-elinating disease. Microb. Pathog. 3:319–326.

2. Caserta, T. M., A. N. Smith, A. D. Gultice, M. A. Reedy, and T. L. Brown.2003. O-VD-OPh, a broad spectrum caspase inhibitor with potent antiapop-totic properties. Apoptosis 8:345–352.

3. Chao, J. R., J.-M. Wang, S.-F. Lee, H.-W. Peng, Y.-H. Lin, C.-H. Chou, J.-C.Li, H.-M. Huang, C.-K. Chou, M.-L. Kuo, J. J.-Y. Yen, and H.-F. Yang-Yen.1998. mcl-1 is an immediate-early gene activated by the granulocyte-mac-rophage colony-stimulating factor (GM-CSF) signaling pathway and is onecomponent of the GM-CSF viability response. Mol. Cell. Biol. 18:4883–4898.

4. Chiu, C., and F. Lee. 1989. IL-6 is a differentiation factor for M1 andWEHI-3B myeloid leukemic cells. J. Immunol. 142:1909–1915.

5. Deszcz, L., J. Seipelt, E. Vassilieva, A. Roetzer, and E. Kuechler. 2004.

Antiviral activity of caspase inhibitors: effect on picornaviral 2A. FEBS Lett.560:51–55.

6. Dzhagalov, I., A. St. John, and Y.-H. He. 2007. The antiapoptotic proteinMcl-1 is essential for the survival of neutrophils but not macrophages. Blood109:1620–1626.

7. Green, D. R. 1998. Mitochondria and apoptosis. Science 281:1309–1312.8. Gross, A., J. Jockel, M. C. Wei, and S. J. Korsmeyer. 1998. Enforced dimer-

ization of BAX results in its translocation, mitochondrial dysfunction andapoptosis. EMBO J. 17:3878–3885.

9. Henning, R., H. R. Stennicke, L. Quinn, E. W. Deveraux, J. C. Reed, V. M.Dixit, and G. S. Salvesen. 1999. Caspase-9 can be activated without proteo-lytic processing. J. Biol. Chem. 274:8359–8362.

10. Jelachich, M. L., P. Bandyopadhyay, K. Blum, and H. L. Lipton. 1995.Theiler’s virus growth in murine macrophage cell lines depends on the stateof differentiation. Virology 209:437–444.

11. Jelachich, M. L., C. Brumlage, and H. L. Lipton. 1999. Differentiation of M1myeloid precursor cells into macrophages results in binding and infection byTheiler’s murine encephalomyelitis virus (TMEV) and apoptosis. J. Virol.73:3227–3235.

12. Jelachich, M. L., and H. L. Lipton. 1999. Restricted Theiler’s murine en-cephalomyelitis virus infection in murine macrophages induces apoptosis.J. Gen. Virol. 80:1701–1705.

13. Jelachich, M. L., and H. L. Lipton. 2001. Theiler’s murine encephalomyelitisvirus induces apoptosis in gamma interferon activated M1 differentiatedmyelomonocytic cells through a mechanism involving tumor necrosis factoralpha (TNF-�) and TNF-�-related apoptosis-inducing ligand. J. Virol. 75:5930–5938.

14. Jelachich, M. L., and H. L. Lipton. 1996. Theiler’s murine encephalomyelitisvirus kills restrictive but not permissive cells by apoptosis. J. Virol. 70:6856–6861.

15. Jelachich, M. L., H. V. Reddi, M. Trottier, B. P. Schlitt, and H. L. Lipton.2004. Susceptibility of peritoneal macrophages to infection by Theiler’s virus.Virus Res. 104:123–127.

16. Kerr, J. F. R., A. H. Wyllie, and A. R. Currie. 1972. Apoptosis: a basicbiological phenomenon with wide-ranging implication in tissue kinetics.Br. J. Cancer 26:239–257.

17. Kirkegaard, K., M. P. Taylor, and W. T. Jackson. 2004. Cellular autophagy:surrender, avoidance and subversion by microorganisms. Nat. Rev. Micro-biol. 2:301–314.

18. Klionsky, D. J., and S. D. Emr. 2000. Autophagy as a regulated pathway ofcellular degradation. Science 290:1717–1721.

19. Kozopas, K. M., T. Yang, H. L. Buchan, P. Zhou, and R. W. Craig. 1993.MCL 1, a gene expressed in programmed myeloid cell differentiation, hassequence similarity to BCL2. Proc. Nat. Acad. Sci., USA 90:3516–3520.

20. Kroemer, G., and S. J. Martin. 2005. Caspase-independent cell death. Nat.Med. 11:725–730.

21. LeBlanc, H. N., and A. Ashkenazi. 2005. Apo2/TRAIL and its death decoyreceptors. Cell Death Differ. 10:66–75.

21a.Lidsky, P. V., L. I. Romanova, M. S. Kolesnikova, M. V. Bardina, S. Hato,F. J. M. van Kuppeveld, and V. I. Agol. 2006. Apoptotic involvement incardiovirus-induced cell death. Europic 2006, abstr. D29.

22. Lipton, H. L. 1975. Theiler’s virus infection in mice: an unusual biphasicdisease process leading to demyelination. Infect. Immun. 11:1147–1155.

22a.Lipton, H. L., G. Twaddle, and M. L. Jelachich. 1995. The predominant virusantigen burden is present in macrophages in Theiler’s murine encephalomy-elitis virus-induced demyelinating disease. J. Virol. 69:2525–2533.

23. Lipton, H. L., and D. H. Gilden. 1996. Viral diseases of the central nervoussystem: persistent infections, p. 853–867. In N. Nathanson (ed.), Viral patho-genesis. Lippincott-Raven, Philadelphia, PA.

24. Lipton, H. L., G. Twaddle, and M. L. Jelachich. 1995. The predominant virusantigen burden is present in macrophages in Theiler’s murine encephalomy-elitis virus (TMEV)-induced demyelinating disease. J. Virol. 69:2525–2533.

25. Liu, H., H. Perlman, L. J. Pagliari, and R. M. Pope. 2001. Constitutivelyactivated Akt-1 is vital for the survival of human monocyte-differentiatedmacrophages: role of Mcl-1, independent of nuclear factor (NF)-�B, Bad, orcaspase activation. J. Exp. Med. 194:113–125.

26. Martinat, C., I. Mena, and M. Brahic. 2002. Theiler’s virus infection ofprimary cultures of bone marrow-derived monocytes/macrophages. J. Virol.76:12823–12833.

27. Mena, I., J. P. Roussarie, and M. Brahic. 2004. Infection of macrophageprimary cultures by persistent and nonpersistent strains of Theiler’s virus:role of capsid and noncapsid viral determinants. J. Virol. 78:13356–13361.

28. Nathanson, N., and F. Gonzalez-Scarano. 2007. Viral persistence, p. 130–145. In N. Nathanson (ed.), Viral pathogenesis and immunity, 2nd ed. Ac-ademic Press, New York, NY.

29. Obuchi, M., Y. Ohara, T. Takegami, T. Murayama, H. Takada, and H.Iizuka. 1997. Theiler’s murine encephalomyelitis virus subgroup strain-spe-cific infection in a murine macrophage-like cell line. J. Virol. 71:729–733.

30. Pena-Rossi, C., M. Delcroix, I. Huitinga, A. McAllister, N. van Rooijen, E.Claassen, and M. Brahic. 1997. Role of macrophages during Theiler’s virusinfection. J. Virol. 71:3336–3340.

31. Ravikumar, B., Z. Berger, C. Vacher, C. J. O’Kane, and D. C. Rubinstein.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

2006. Rapamycin pre-treatment protects against apoptosis. Hum. Mol.Genet. 15:1209–1216.

32. Riedl, S. J., and G. S. Solvesen. 2007. The apoptosome: signalling platformof cell death. Mol. Cell. Biol. 8:405–413.

33. Rodriguez, J., and Y. Lazebnik. 1999. Caspase-9 and APAF-1 form an activeholoenzyme. Genes Dev. 13:3179–3184.

34. Rozhon, E. J., J. D. Kratochvil, and H. L. Lipton. 1983. Analysis of geneticvariation in Theiler’s virus during persistent infection in the mouse centralnervous system. Virology 128:16–32.

35. Ruhl, S., and D. H. Pluznik. 1993. Dissociation of early and late markers ofmurine myeloid differentiation by interferon-gamma and interleukin-6.J. Cell Physiol. 155:130–138.

36. Rust, R. C., L. Landmann, R. Gosert, B. L. Tang, W. Hong, H.-P. Hauri, D.Egger, and K. Bienz. 2001. Cellular COPII proteins are involved in produc-tion of the vesicles that form the poliovirus replication complex. J. Virol.75:9808–9818.

37. Steurbaut, S., B. Rombaut, and R. Vrijsen. 2007. Theiler’s virus-strain de-pendent induction of innate immune responses in RAW264.7 macrophages

and its influence on viral clearance versus viral persistence. J. Neurovirol.13:47–55.

37a.Vandenabeele, P., T. Vanden Berghe, and N. Festjens. 2006. Caspase inhib-itors promote alternative cell death pathways. Sci. STKE 358:pe44.

38. Wileman, T. 2006. Aggresomes and autophagy generate sites for virus rep-lication. Science 312:875–878.

39. Willis, S. N., J. I. Fletcher, T. Kaufmann, M. F. van Delft, L. Chen, P. E.Czabotar, H. Ierino, E. F. Lee, W. G. Fairlie, P. Bouillet, A. Strasser, R. M.Kluck, J. M. Adams, and D. C. S. Huang. 2007. Apoptosis initiated whenBH3 ligands engage multiple Bcl-2 homologs, not Bax or Bak. Science315:856–859.

40. Wolter, K. G., Y.-T. Hsu, C. L. Smith, A. Nechushtan, X.-G. Xi, and R. J.Youle. 1997. Movement of Bax from the cytosol to mitochondria duringapoptosis. J. Cell Biol. 139:1281–1292.

41. Zhou, P., L. Quian, K. M. Kozopas, and R. W. Craig. 1997. Mcl-1, a Bcl-2family member, delays the death of hematopoietic cells under a variety ofapoptosis-inducing conditions. Blood 89:630–643.



http://jvi.asm.org/

Dow

nloaded from

http://jvi.asm.org/

Theiler's Virus-Induced Intrinsic Apoptosis in M1-D Macrophages Is ...

Documents

Transcript of Theiler's Virus-Induced Intrinsic Apoptosis in M1-D Macrophages Is ...